The main aim of this paper was to determine the optimal design of a sand mould using computer simulation and statistical optimisation techniques. Simulations of the sand-casting process were performed using NovaFlow&Solid software. The geometry of the mould and the type of sand used were varied according to the Taguchi L18 design of experiments. Output variables used to assess process quality and efficiency were total casting time, volume shrinkage, and melt efficiency. Taguchi-based grey relational analysis was used to identify the optimal process parameters. The combination of input parameters represents the optimal solution for process simulation resulting in minimal total casting time, minimal volume shrinkage and maximum melt efficiency. The proposed methodology provides an effective framework for simplifying the casting process and enhancing the reliability of gravity-cast components.